DNA polymerases catalyze the synthesis of DNA and can be found in all cells as well as being encoded in numerous viruses. Although all DNA polymerases possess 5′-3′ DNA polymerization activity, DNA polymerases differ from one another in numerous other properties. For example, some enzymatic activities that are possessed by some DNA polymerases, but absent in other DNA polymerases include: double stranded DNA 5′-3′ exonuclease activity, single-stranded DNA 3′-5′ exonuclease activity, double-stranded 3′-5′ DNA exonuclease activity, RNase H activity, reverse transcriptase activity, and the like. Additionally, different DNA polymerases may have different optimal pH and temperature ranges for activity. Furthermore, DNA polymerases may differ in the rate in which they catalyze DNA synthesis.
Purified DNA polymerases have numerous uses in vitro. A detailed description of DNA polymerases, including methods for their isolation, can be found among other places, in DNA Replication 2nd edition, by Komberg and Baker, W.H. Freeman & Company, New York, N.Y. 1991. In vitro uses of DNA polymerases include, for example, the labeling and synthesis of hybridization probes, DNA sequencing, and DNA amplification. A DNA amplification method employing DNA polymerases that has been particularly useful is the polymerase chain reaction (PCR) technique which employs the use of a thermostable DNA polymerase.
The first thermostable DNA polymerase that is widely used for DNA amplification is Taq DNA polymerase isolated from the thermostable, aerobic bacterium Thermus aquaticus. Taq DNA polymerase's enzymatic activity at high temperatures allows for primer extension and sequencing of polynucleotide templates with complex secondary structures (i.e., by PCR amplification). However, Taq DNA polymerase has significant error rate when incorporating nucleotides due to the lack of 3′-5′ exonuclease activity (i.e., proofreading activity), and therefore may not be suitable if the amplified sequence is to be used in further gene structural/functional studies or cloning.
In the last 10 years, numerous studies have quantified the error rate of thermostable DNA polymerases, and several enzymes have been found to copy DNA more accurately than Taq DNA polymerase (referred to as high fidelity DNA polymerases). U.S. patent describing DNA polymerases include U.S. Pat. Nos. 4,492,130; 4,946,786; 5,210,036; 5,420,029; 5,489,523; 5,506,137; 5,545,552; 5,618,711; 5,624,833; 6,238,905; 6,100,078; 6,077,664; 5,968,799; 5,948,663; 5,885,713; 5,834,285; 5,756,334; 5,747,298; 5,744,312; 5,624,833; 5,602,011; 5,556,772.
High fidelity polymerases alone should definitely increase fidelity rates but usually do not amplify long fragments as efficient as a DNA polymerase lacking a 3′-5′ exonuclease activity (e.g., Taq DNA polymerase). Enzyme mixtures that combine a standard polymerase with a small amount of proofreading polymerase may provide a balance between fidelity and yield. A study published in 1994 illustrated that the use of a high level of a DNA polymerase lacking 3′-5′ exonuclease activity (an exo− DNA polymerase, Klentaq-1) with a very low level of a thermostable DNA polymerase exhibiting 3′-5′ exonuclease activity (an exo+ DNA polymerase such as Pfu, Vent, or Deep Vent) generated products with increased base-pair fidelity with a maximum yield of 35 kb DNA from 1 ng of lambda DNA template (Barnes, Proceedings of the National Academy of Sciences, 91:2216-20, 1994). Similarly, U.S. Pat. Nos. 5,436,149 and 6,008,025 disclosed methods for improving DNA amplification fidelity using a DNA polymerase composition comprising a first enzyme substantially lacking 3′-5′ exonuclease activity and a second enzyme comprising 3′-5′ exonuclease activity. In mixtures such as these, the exo+ enzyme acts to correct polymerization errors produced by the exo− DNA polymerase.
The problem inherited in the above composition comprising the mix of two DNA polymerases is that the high polymerization activity resulted from combining the two DNA polymerases may inhibit the efficiency and therefore the yield of the amplification reaction. Therefore, one can not increase fidelity by increasing the proportion of the proofreading DNA polymerase without compromising PCR product yield. It is also known that the amplification fidelity may also be affected by high DNA polymerase concentration (see for example, Mattila et al., 1991, Polynucleotides Research, 19:4967-73).
There is therefore a need in the art for new methods and compositions which improve polymerization fidelity and reduce the side effects resulted from having high polymerization activity in the reaction.